View
0
Download
0
Category
Preview:
Citation preview
1
DO: 10.1002/chem.200((will be filled in by the editorial staff))
Directed Remote Aromatic Metalations: Mechanisms and
Driving Forces
David Tilly*,[a]
Jakob Magolan,[b]
Jacques Mortier[c]
2
Introduction
Directed metalation has become a powerful tool for regioselective
access to elaborate polysubstitued aromatic and heteroaromatic
scaffolds. Gschwend, Rodriguez,[1] Snieckus,[2] Beak,[2b, 3]
Schlosser[4] and Clayden[5] have effectively illuminated the broad
scope and utility of directed metalation strategies in synthesis. In
this review, we focus our discussion on the rationalization of
regioselectivity of metalations at remote positions relative to the
directing species or directed remote metalations (DreM, Figure 1).
Directed remote aromatic metalations have served as key
transformations in the synthesis of natural products, active
pharmaceutical ingredients and materials.[2b] The mechanistic
concept of complex induced proximity effect (CIPE), originally
introduced in association with directed ortho metalations (DoM),
was recently also utilized to rationalize DreM reactivity.[2b]
However, a number of recent mechanistic studies of DreM have
provided a deeper understanding of these transformations that
extend beyond the scope of the CIPE model.
Figure 1. Directed ortho metalation (DoM) and directed remote metalation (DreM).
In this review, we strive to unravel the tangle of data reported
on DreM using a mechanistic framework. Metalation
regioselectivities are governed by a combination of kinetic and
thermodynamic factors that are highlighted throughout the review.
Through careful analysis of critically selected synthetic examples
and mechanistic investigations, we offer insights intended to aid
chemists in the prediction and rationalization of regioselectivity of
directed metalations.
The mechanistic concepts applicable to directed metalations
(CIPE, kinetically enhanced metalations, and overriding base
mechanisms) will first be succinctly presented and then critically
evaluated throughout the review for a series of substrate classes.
Reaction mechanisms, and the consequent regioselectivity
outcomes, depend on numerous factors including: the rigidity of
the reactant skeletons, the nature and strength of the base, the
coordinating, electrophilic and migrating properties of directing
metalation groups (DMG), the stability of metalated products, the
nature of the external electrophiles, the acidity of aromatic
hydrogens, the kinetics of reactions among other parameters.
DreM reactions are classified here into four substrate categories
and each is considered in sequence (Figure 2, A-D). Firstly,
DreM-intramolecular quench sequences (Figure 2, A and B) are
discussed and significant driving forces are evaluated. In this
category, directed remote aromatic and lateral metalation (B) on
flexible biaryl structures bearing an electrophilic coordinating
DMG is followed by intramolecular quench of remote anion with
the DMG. Isomerization of the remote kinetic anion to the
thermodynamically more stable species prior to electrophilic
intramolecular quench is sometimes observed and translocation of
a migrating DMG is also possible as part of the quench step.
Secondly, remote metalations that are quenched by external
electrophiles are analyzed (Figure 2, C and D). Trapping of
remote anions with external electrophiles has been reported for
non-electrophilic DMGs and for electrophilic coordinating DMGs
when the rigidity of structures does not allow for subsequent
intramolecular quench.
Lithiation on the meta and/or para positions of an arene with
respect to the directing group,[6] perilithiations,[7] Parham
cyclizations[8] have been recently reviewed and are not discussed
here.
Abstract: Directed remote aromatic metalations are useful synthetic transformations allowing for rapid regioselective access to elaborate highly substituted carbocyclic aromatic and heteroaromatic systems. This review unravels the tangle of data reported on directed remote aromatic metalations. Through a careful analysis of critically selected examples, advanced rationalizations of remote metalation regioselectivities are presented. These extend beyond the complex induced proximity effect (CIPE).
Mechanisms, driving forces and parameters influencing remote metalations are discussed. An understanding of these metalation mechanisms enables more accurate predictability of justification of regiochemical outcomes of these useful synthetic transformations. Keywords: directed metalation • remote metalation • lithiation • regioselectivity • reaction mechanisms
[a]* Dr. D. Tilly Eskitis Institute for Cell and Molecular Therapies Griffith University Don Young Road, Nathan 4111, Queensland, Australia E-mail: dp.tilly@laposte.net
[b] Dr. J. Magolan Department of Chemistry, University of Idaho Moscow, ID, USA
[c] Prof. J. Mortier University of Le Mans UCO2M (UMR CNRS 6011) Avenue Olivier Messiaen, 72085 Le Mans, Cedex 9, France
3
Figure 2. Directed ortho metalation (DoM) and directed remote metalation (DreM).
Mechanistic Concepts in Directed Metalation
Metalation by organometallic bases in the vicinity of heteroatom-
containing functional groups often are rapid, regioselective and
efficient.[1-6] The specific mechanistic pathways that allow the
directing metalation group (DMG) to enhance the rate and
regioselectivity of metalation within its vicinity remain the subject
of debate.[2b, 9] Mechanism concepts that are commonly invoked
to rationalize metalation regioselectivities are presented below.
Complex induced proximity effect (CIPE): The CIPE concept
was introduced in organolithium transformations to rationalize the
formation of kinetic rather than the anticipated thermodynamic
products and the acceleration of seemingly unfavorable
reactions.[1-3] According to the CIPE model, a pre-lithiation event
brings reactive groups into close proximity before the occurrence
of subsequent intramolecular rate-determining metalation at the
proximate site (Scheme 1). Over the years, the CIPE concept has
provided guidance for synthetic development and rationalization
of regioselective metalations including directed remote metalation
(DreM). However, CIPE is commonly invoked without direct
evidence of complexation to confirm “directed” metalations.
CIPE was recently described by Snieckus and Beak as an efficient
predicative tool useful as “heuristic model” regardless of the
detailed reaction mechanism.[2b]
The involvement of CIPE in metalation should not be
considered a general concept but rather requires investigation of
the specific reaction and the nature of the substrate. In most cases,
experimental data are subject to interpretation and a definite
mechanism cannot be stated.[2b, 9b] The literature offers abundant
evidence for ligation of lithium to heteroatoms in ground states.[10]
The formation of pre-lithiation complexes between the
organoalkali base and the reactant on the reaction pathway to
directed metalation is supported by kinetic data,[11]
computational[2b] and IR data[12] in several transformations.[13] A
CIPE, or lack thereof, can be manifested by divergent reactions.
Scheme 1. CIPE in deprotonation of arenes (Beak and Meyers, 1986)[3b]
Kinetically enhanced metalation: Complex formation before rate
limiting transition step is often presented as critical to pre-
organizing the species undergoing metalation. It is important to
note that NMR, IR and X-ray studies can detect stable complexes
which may not be on the reaction pathway.[14] Any number of
transiently formed species can precede the rate limiting step with
none being kinetically relevant.[11d, 15] Schleyer has postulated a
model in which complexation of the base and proton abstraction
occur simultaneously.[16] According to this model, directing and
accelerating effects of the DMG are attributed to the existence of a
stabilizing metal-substituent interaction at the rate determining
transition structure rather than to a pre-lithiation complex. The
term “kinetically enhanced metalation” was coined as an
alternative to the ‘complex-induced proximity effect’ terminology.
The kinetically enhanced metalation model focuses both on
agostic interactions between lithium and the ortho hydrogen atoms
lowering the energy at the transition state, and on favorable charge
distribution in the transition structures (Figure 3).[16]
Figure 3. Four-center transition structure for DoM (Hommes and Schleyer, 1992).[16]
David Tilly graduated as Engineer from Ecole
Nationale Supérieure de Chimie Rennes, and with a
Master of Chemistry from Université de Rennes
(France). In 2005, he completed a PhD in Organic
Chemistry at Université du Maine (France) with Prof.
J. Mortier (Directed metalations). After postdoctoral
positions tackling the syntheses of complex diterpenes
(Brisbane) and palladium catalysis (with Prof. V.
Gevorgyan, University of Illinois, Chicago), he works
since 2008 at ESKITIS (Australia) with Assoc. Prof. M.
Coster, developping potent anticancer drugs.
Jakob Magolan received his BSc from Queen’s
University and PhD from the University of Western
Ontario in Canada. His doctoral studies were in the
field of heterocyclic natural products synthesis under
the guidance of Prof. M. Kerr. After a postdoctoral
fellowship at Griffith University, Australia, he began
his independent academic career in 2010 at the
University of Idaho. Dr. Magolan’s research
interests include: heterocyclic synthetic methods,
heterogeneous catalysis, one-pot multi-step
methodology, and anti-cancer medicinal chemistry.
Jacques Mortier obtained his PhD at Université de
Rennes 1 (France) with Prof. J. Hamelin. After a
postdoctoral position at Northwestern University,
USA (with Prof A. G. M. Barrett, 1987), he was
appointed for 6 years as a senior scientist at Rhône-
Poulenc, Agrochemical Research Centres of Lyons,
France and Ongar, England. In 1994, he was
appointed as Assistant Professor in Rennes, obtained
his Habilitation in 1995, and was nominated member
of Institut Universitaire de France in 1998. Since
1999, he is Professor at Université du Maine (Le
Mans, France).
4
In our opinion, both theories (CIPE and kinetically enhanced
metalation) furnish a complete picture if adequately combined.[9a]
Given the abundant literature evidence for coordination of lithium
to heteroatoms in ground states (vide supra), it is very probable
that initial approach of the base to the substrate is mediated by
strong coordination with the electron-rich π-system of the DMG,
leading to a pre-lithiation complex PLC (kinetic control, Figure 4).
The directed ortho-lithiation was suggested to proceed by a two-
step mechanism in which initial complexation is reversible
whereas second step is rate determining.[9b] Not only must the
reactants be brought together by chelation of the base with the
DMG to form a pre-lithiation complex (kinetic control, Figure 4),
they have to be held in exactly the right orientation relative to each
other in the transition states to ensure that deprotonation can occur.
Both of these factors raise the free energy of the system by
lowering the entropy. Some energy also must be invested to begin
breaking the C-H bond so that the C-Li bond can form. The
directing and accelerating effect of substituents is likely due to the
stabilization of both the initial complex and the transition structure.
The metal is involved in partial bonds and coordination by the
DMG becomes stronger in the transition state than in the initial
complex. As a result, complexation increases the rate of reaction
by providing a new mechanism with lower activation energy (Ea).
There are two transition states, each with its own activation energy
(Ea1 and Ea2). The overall activation energy is the difference in
energy between the reactant state and the highest energy transition
state (Ea). Notably, the geometries of precursor complex and
transition structure can be radically different.
Figure 4. Qualitative energy diagram for directed metalations.[9a]
The relative basicities of n-BuLi in cyclohexane as a function
of the addition of increasing increments of THF or TMEDA were
recently assessed by kinetics studies and 7Li NMR studies.[17] A
gradual controlled increase in the basicity of n-BuLi was observed
with increasing increments of THF or TMEDA. Relationships of
the basicity in the varied media to the three oligomeric forms of n-
BuLi were proposed. Complexations of metal bases to Lewis
basic DMGs may likely have a similar effect on the relative basic
properties intra-aggregate.
Overriding base mechanism: The overriding base mechanism
suggests that some strong bases react preferentially with relatively
acidic aromatic protons without pre-coordination to a substituent.[1,
11d] The mechanism depends solely upon the enhanced acidity of
the ortho hydrogens resulting from strong electronegativities of
certain DMGs such as halo, trifluoromethyl, and cyano moieties.
Inductive effects were advantageously exploited in the metalation
of arenes bearing two heteroatom-based DMGs. Weakly solvated
organolithium reagents preferentially exploit the coordinative
ability of a DMG whereas fully complexed bases (with t-BuOK or
PMDTA) do not complex weaker coordinating functional
groups/ligands but rather selectively target positions where a
negative charge can be most efficiently stabilized blah[4, 18]
DreM: Early Mechanistic Interpretations
Directed remote metalation methodology is often presented as an
extension of directed ortho metalation whereby the directing
ability of DMG apparently reaches beyond the ortho
hydrogens.[2b] The labels ‘directed remote’ metalation or ‘through-
space’ induced metalation describe metalations assisted by a
substituent at position that is ‘formally’ remote from, but
conformationally in proximity to a functional group that is likely
to be complexed by a metalating agent (Figure 2).[2b] The study of
DreM has occurred at a slower pace that of DoM as positions
ortho to directing groups appear to be more readily metalable as
dictated by CIPE.[2]
Early examples of directed remote aromatic metalations were
reported on biarylic structures bearing amino groups (vide
supra).[19] Cartoon and Cheeseman also reported the directed
lithiation of 1-(2’-carboxyphenyl)pyrrole, 1, with LDA to form 9-
keto-9H-pyrolo[1,2-a]indole, 3, via the “anionic equivalent of the
Friedel-Crafts cycloacylation.’ (Scheme 2).[20] The authors
suggested the metalation was “directed by the possibly chelated
lithium cluster present on the adjacent ring and by the relative
acidity of pyrrole hydrogens.”[20-21] ] DreM with nucleophilic
intramolecular quench of biarylcarboxamides was reported by
Snieckus to afford exclusively fluorenones.[22]
Scheme 2. Early example of DreM with cyclization directed by carboxyl DMG
(Cartoon and Cheeseman, 1981)[20]
DreM methodology has been developed, primarily by
Snieckus and co-workers, for a variety of directing groups and
substrates.[23] The versatility and practicality of DreM in synthesis
was established by a seminal work of Snieckus: DreM proved very
dependable without extensive mechanistic insights and acted as a
conduit for the synthesis of numerous natural products including
dengibsinin,[22] 6-deoxydengibsin,[24] dengibsin,[25] lmeluteine,[23b]
defucogilvocarin,[26] piperolactam,[27] β-lapachone,[28] plicadin,[29]
eupolauramine,[30] gymnopusin C,[31] triazadibenzo[cd,f]azulen-
7(6H0-one,[32] salcomine[33] and homoschatoline[33] (Figure 5).
5
Figure 5. Examplex of complex compounds accessed using DreM-cyclization.
Notably, most DreM reactions have been reported on flexible
biaryl structures bearing electrophilic DMG yielding products that
result from intramolecular quench of remote anions by the DMG
only (Figure 2, A and B). There are few reported examples of
quench of a remote anion by external electrophiles. These contain
exclusively non-electrophilic DMGs or rigid substrate scaffolds
that do not allow for intramolecular quench (Figure 2, C and D).
The apparent selective formation of products from intramolecular
quench of remote anions was initially interpreted as indirect
evidence for efficient regioselective remote metalation first steps. [22-33] The remote regioselectivity of metalation was rationalized
initially through the involvement of CIPE only.[2b] As illustrated
in Figure 6, a DMG2 functionality was proposed to convert the
‘remote’ site into a hypothetical ‘ortho’ site for metalation on the
adjacent ring.
Figure 6. CIPE used to explain DreM on 2-DMG-biphenyl substrates.
This rationalization is problematic as CIPE is already used to
account for ortho metalation regioselectivities on biaryl
substrates.[34] Given that the hydrogens in the ortho position
relative to the heteroatom containing DMG1 (Figure 6) are more
inductively activated and closer to the Lewis basic group than
remote hydrogens, it is inadequate to use CIPE alone to rationalize
to rationalize both regioselectivities (ortho and remote) on biarylic
reactants? Moreover, experimental conditions that favor CIPE
(alkyllithium reagents, low temperatures) are known to lead to
DoM[34] whereas conditions that are likely to disfavor coordination
to DMG (use of amide bases and ligands such as HMPA, tBuOK,
higher temperature) preferentially lead to DreM. [22-33, 35] In the
following sections, we aim to clarify the driving forces of remote
aromatic metalations by highlighting relevant recent studies. The
various DreM mechanisms are classified according to structural
features of the reactants.
DreM with Intramolecular Quench by Electrophilic DMG
Directed remote metalations on flexible biarylic structures bearing
non migrating electrophilic DMGs such as carbamide, carboxylate,
or ester moieties form cyclic compounds as sole products. [22-33, 35a]
To date, attempts to trap remote metalated species on these
systems with external electrophiles has failed. Cyclized products
result from intramolecular nucleophilic substitutions of remote
anionic species onto the electrophilic DMG (Figure 2, A and B).
With an electrophilic migrating DMG such as a carbamate,
intramolecular quench of preformed remote anions is possible on
both rigid or flexible substrate skeletons. In the following sections,
the various mechanisms of DreM with intramolecular quench are
unfolded. Driving forces are highlighted and the parameters
allowing for efficient remote metalations are discussed.
In Situ Quench Mechanisms and Influence of the Nature of the
Base
Mechanisms of DreM-cyclizations were recently investigated on
2-biphenylcarboxylic acid[35a] and 2-biphenylamide.[36] These
representative substrates were chosen for mechanistic studies
since their scaffolds are prototypes of DreM-intramolecular
quench methodology, thus mechanistic information gathered on
DreM-cyclization of such structures can potentially be extended to
more complex analogous reactants. These fundamental studies are
presented below. In contrast to initial assumptions, DreM
reactions were found to be non-regioselective.[35-36] That is,
metalation at the ortho position is kinetically favored (via CIPE)
even when cyclized compounds from DreM are the only products
isolated. DMGs act as intramolecular electrophiles in a similar
manner to in situ quench (ISQ) techniques.[37] The flexibility of
the reactant structures or/and the migratory ability of electrophilic
substituents (i.e. carbamates) allow for rapid intramolecular
quench of remote anions. Furthermore, DreM mechanisms are
found to be base-dependent.
DreM-intramolecular quench under irreversible metalation
conditions. Example of 2-biphenylcarboxylic acid and
LiCKOR: The first thorough mechanistic study of directed
remote aromatic metalation was reported on 2-biphenylcarboxylic
acid (Scheme 3).[35a] Substrate 4 undergoes metalation in the
immediate vicinity of the carboxylate substituent (C3) when
treated with s-butyllithium in THF.[34c, 34d] In contrast, treatment
of 4 with the LICKOR superbase (n-butyllithium/tBuOK)[38]
affords the fluorenone 7 (76% in benzene at 60 °C) after
hydrolysis. This appears to be a regioselective remote metalation.
Furthermore, the use of amide bases did not result in remote
metalation of 4, presumably because the carboxylate moiety does
not activate remote positions through “through-bond” electronic
activation to a sufficient degree. A stronger base is a requirement.
Importantly, LiCKOR base irreversibly metalates reactants.
6
Scheme 3. Optional site selectivity in metalation on 2-biphenylcarboxylic acid.[34c, 35a]
Isotope labelling experiments established that LiCKOR
metalates non-regioselectively in both the ortho (C3) and remote
(C2’) positions of 4 (Scheme 4). The ortho lithiated species 5
(obtained by treatment of 4 with sBuLi) is stable at room
temperature and does not equilibrate with the remote anionic
species 8. Following metalation at remote (C2’) positions the
carboxylate moiety 8 reacts with the remote C2’ aryl anion via fast
and irreversible nucleophilic cyclization to form a stable
tetrahedral gem-dialkoxide, intermediate 6, prior to hydrolysis.[35a]
This nucleophilic cyclization ultimately drives the reaction
towards the formation of fluorenones exclusively. No measurable
build-up of a remote anionic species 8 indicates a rate-limiting
remote lithiation followed by a rapid annulation. Strong support
for a fast and irreversible cyclization step is found in the literature.
When 2’-bromo-2-biphenylcarboxylic acid, 2’-bromo-2-
biphenyloxazoline,[39] and 1-(2’-bromophenyl)pyrrole-2-
carboxylic acid[20] were subjected to halogen-lithium exchange,
the 2’-lithiated species could not be trapped by an external
electrophile in any case. Within the past decades, reported
attempts to trap an anion remote from electrophilic DMGs during
DreM-cyclization sequences have repeatedly failed. [22-33, 35]
Ortho metalated species 5 also leads to the cyclized structure 9
upon exposure to LiCKOR. Notably, directing metalation
properties were also found for gem-dimetalo dialkoxide
[C(OM)2].[35a]
Scheme 4. Directed remote metalation of 2-biphenylcarboxylic acid (4) with
LiCKOR.[35a]
In the above example, the involvement of CIPE is unclear and
an overriding mechanism being possible given that LiCKOR is
able to efficiently metalate naked benzene rings.[38] The
carboxylate moiety might also compete with tBuOK as a solvation
ligand. In such case, mixed aggregates would generate agostic
hydrogen-metal interactions in the intermediate complexes[40] and
delivery of the base by prior coordination to the carboxylate would
assist metalation in a non-regiospecific fashion. These results
neither require nor exclude the existence of pre-equilibrium
complexes and aggregates. Whereas 5 (M = Li) is stable at room
temperature, a fast Li-K permutation may allow the equilibrium
between 5 and 8 to be effective.[41]
DreM-nucleophilic cyclization under reversible metalation
conditions. Example of N,N-dialkylbiphenyl-2-carboxamide
and amide base: Another example of optional site selectivity was
reported on N,N-diethyl-2-biphenylcarboxamide 12 (Scheme 5).
Metalation of 12 with sBuLi/TMEDA in THF at -78 °C occurs at
the ortho position (DoM, via CIPE).[34a, 34b] In contrast, subjection
of 12 to LDA at 0 °C to RT yields fluorenone 7 (DreM).[31, 35a] In
this case, the mild amide base reversibly metalates the substrate
allowing for series of equilibria between lithiated species during
the reaction.[42]
Scheme 5. Metalations of 2-biphenylcarboxamide (12) with lithium bases.[22, 34a, 34b, 36]
Isotope labelling experiments indicated that LDA metalation
of 12 is non-regioselective,[36] with metalation at the ortho
position being favored in accordance with the CIPE model. When
LDA and TMSCl were premixed in THF solution (-78 °C) prior to
addition of 12 (ISQ), 14 was formed exclusively (65%, Scheme 6).
This suggests kinetic control predominates over thermodynamic
acidity of C-3H compared to C-2′H in 12. Ortho-lithiated species
13, formed by reaction of 12 with sBuLi/TMEDA (2.2 equiv.) at -
78 °C, is stable when allowed to warm to room temperature and
does not equilibrate with 15 in the absence of some other
contributor. However, addition of 0.1 equiv. of diisopropylamine
to 13 followed by hydrolysis yields fluorenone 7 in 30 % yield.
A mechanism for DreM-cyclization of 12 was proposed based
on experimental data (e.g. isotope labelling) (Scheme 6).[39] It
involves an initial amide-base complexation and equilibrium
formation of complex C, which reacts fast with an in situ
electrophile (Me3SiCl) to afford 14. This rapid reaction prevents
equilibration of 12 with 15 at temperatures below -30 °C. Above -
30 °C however, when generated with an amide base, 13 does
undergo undergoes equilibration via 12 to 15 (likely formed in
traces amount only) which rapidly cyclizes to a stable tetrahedral
carbinolamine oxide 16. Upon hydrolysis 16 affords fluorenone.
The fast irreversible nucleophilic cyclization step drives
equilibrations towards the formation of fluorenone-precursor 16
only. The flexibility of the substrate structure, allowing for
intramolecular cyclization, is an important feature that permits
DreM reactivity. The driving forces for remote metalation of 12
(ISQ) are potentially significantly different from the ones
described for substrate 4 (Schemes 3 and 4). The nature of the
base, either reversibly or irreversibly metalating reactants, leads to
alterations of mechanisms and reaction condition requirements.
7
Scheme 6. DreM-nucleophilic cyclization of 2-biphenylcarboxamide (12) with LDA
(Snieckus and Mortier, 2010) [36]
Strong indirect experimental evidence suggests the
carbinolamine oxide 16 is the stable organometallic species prior
to hydrolysis.[36] Corroborative support for carbinol amine
intermediates most likely existing as dimers or larger complexed
species is inferred by a qualitative IR spectrometric study on
DreM of a related biaryl 2-amide (Scheme 7).[43] Treatment of
N,N-diethyl-4-(3-methoxy-phenyl)-nicotinamide 17 with LDA
leads to the gradual disappearance of the amide carbonyl
stretching frequency (ν = 1632 cm-1) of the starting amide 17 as a
function of time upon gradual addition of 3 equiv. of LDA. After
addition of MeOH, the stretching absorption of the carbonyl (ν =
1718 cm-1), representing azafluorenone 19, increases in intensity.
Scheme 7. React IR spectrometric profile of DreM-cyclization of 17 with LDA
(Snieckus et al., 2007) [43]
Reaction Paramaters for DreM with Intramolecular Quench
The influence of coordination properties and steric hindrance at
the DMG, conformation constraints in the molecule, acidity of
hydrogens at positions remote from DMG, and temperature in
DreM transformations are discussed below. An analysis of these
variables offers guidance for the selection of reaction conditions
for optimum DreM efficiencies.
Activating properties of DMG: Various electrophilic DMGs on
flexible biaryl substrates are suitable for efficient DreM-
intramolecular quench sequences. The efficiency of a DMG to
promote remote metalation relies on an intimate combination of
electrophilic, Lewis basic, and acidifying properties along with
suitable the conformation of the molecule.
Strong coordinating properties of a directing group do not
guarantee effective remote metalation directing suitability. For
example, 2-biphenylcarboxamides and 2-biphenylcarbamates
undergo efficient DreM under mild conditions when mixed with
LDA. [23-33] In contrast, 2-biphenyloxazoline fails to metalate at
the remote position using LDA with the oxazoline moiety being a
good Lewis base but weaker electrophile. The carboxylate group
is a better electrophile but weaker Lewis base than the amide
group however 2-biphenylcarboxylic acid does not cyclize under
LDA conditions.[35a] Carbamates and amides show better
directing group efficiencies towards remote metalation under mild
conditions presumably by achieving better remote activation
through formation of complexes and long range acidification
effects as already observed in DoM.[1-2, 3b]
Steric hindrance at the DMG: Steric hindrance at the DMG is
detrimental to the efficiency of DreM-intramolecular quench.[22, 44]
Steric hindrance may lead to conformational constraints which
cause the coordinating electrophilic DMGs to be distal from the
remote position,[44b] thus disfavoring any CIPE and making
intramolecular electrophilic quench difficult. To date, a remote
anion has yet to be trapped by an external electrophile when
intramolecular substitutions are disfavored by steric hindrance.
Such observation is consistent with mechanisms where remote
metalations at traces amount are followed with intramolecular
irreversible rapid quench driving equilibria towards the efficient
formation of products.
Conformational constraints: The importance of conformational
constraints was highlighted in a synthesis of kinobscurinone,[45]
which involved a mixture of two stable atropisomers 20 A/B
separable by chromatography (Figure 6). When the isomeric (1:1)
A/B mixture was treated with LDA (6 equiv.) at room temperature,
only A was converted into the cyclized product. Competitive
unknown side reaction(s) precluded a more efficient conversion of
B into fluorenone by A/B equilibration. At lower temperatures the
reaction did not proceed at significant rates while at 45 °C poor
yields (38%) of fluorenone were obtained with no recovery of 20.
The authors invoked the CIPE model and cited restricted aryl-
amide bond rotation. The isomer A-LDA complex leads to
productive metalation and cyclization while isomer B-complex
cannot achieve the transition state for analogous deprotonation.
The rate of equilibration of B or its LDA complex with the
equivalent species of isomer A must then be slow relative to their
decomposition pathways, thus precluding higher conversion into
the cyclized product.
As an alternative explanation, we propose that a remote
anionic species of B, formed in traces amount only, could fail to
efficiently cyclize due to conformational constraints, thus
depriving the DreM-cyclization process from its in situ quench
driving force.
Figure 7. Atropisomers intermediates in the synthesis of kinobscurinone (Snieckus et
al., 1997) [45]
Acidity of remote hydrogens: The thermodynamic acidities of
remote hydrogens influence the regioselectivity of DreM. Remote
hydrogens can be acidified via inductive effects and by
coordination of metal bases with proximal substituents.[22, 46]
8
Acidified remote positions are metalated preferentially, affording
regioselective transformations often complementary those
observed for Friedel-Crafts reactivity. [22-33] Heteroatom bridged
arenes, where an atom or a group of atoms (SO2,[47] O,[23i, 44a]
P(O)Ar,[48] NR,[23a, 23h, 44] CO,[49] ) separates the two aromatic rings,
have served as suitable metalation substrates for the synthesis of
many heterocyclic structures.[50] To illustrate thermodynamic
acidity effects in DreM, 6-chloro-3’-methoxy-2-
biphenylcarboxylic acid 21 gives 1-methoxy-5-chlorofluoren-9-
one 22 as the sole product upon treatment with LDA (Scheme
8).[51] In contrast, 6-chloro-2-biphenylcarboxylic acid 23 does not
cyclize under these conditions. A stronger base, LiCKOR, in
benzene at 60 °C is necessary to afford 5-chlorofluoren-9-one 24
(39%) via metalation at C2’.
Scheme 8. DreM-cyclization on 6-chloro-2-biphenylcarboxylic acids (Mortier et al.,
2008) [50]
In Scheme 9, azabiaryls were treated with LDA to form
azafluorenones.[43] These products result from metalation on the
phenyl ring rather than the more acidic pyridine ring. Small
equilibrium concentrations of the lithiated species at C2‘ and C2
are generated, the reaction is driven by an irreversible
intramolecular nucleophilic cyclization and by the activating
effects of remote substituents.
Scheme 9. DreM-cyclization to azafluorenones (Snieckus et al., 2007)[43]
The presence of heteroatoms in the substrates offers further
opportunities for alternate metalation sites. Lower yields of
DreM-nucleophilic cyclization are obtained when substituents
cause side reactions and tuning of reaction conditions is
required.[47-48]
Temperature: Most DreM with intramolecular quench reactions
take place at temperatures above -50 °C. [20-33, 35-36, 44, 52] So far,
attempts to trap remote anions at lower temperatures have failed
with starting materials recovered. This observation is of
mechanistic significance as lower temperatures do not allow for a
fast intramolecular irreversible nucleophilic quench and therefore
DreM does not happen at low temperatures (~-78 °C). DreM-
annulations directed by amide or ester groups are efficiently
achieved at temperatures above -30 °C when electrophilic
substitutions become efficient. Some DreM-anionic remote Fries
rearrangements even require heating.[25, 45, 47-48, 53]
The studies on metalation of N,N-diisopropyl-2-(4-chloro-2-
pyridyl)benzamide 25 and derivatives (Scheme 10) highlight the
influence of temperature on the kinetics of nucleophilic
cyclization.[54] At -75 °C, LTMP in THF promotes the
regioselective metalation of 25 at C5’ only. This is a position far
from the carbonyl function but proximal to the chlorine substituent.
Deuterium incorporation occurs quantitatively at C5‘ only.
Despite strong coordinating ability of the carbonyl DMG, the
metalation is directed by the chlorine substituent which acidifies
the hydrogens at C3’ and C5’ of 25 and exerts a stabilizing effect
on the corresponding lithio-derivative. Position C5’ was
determined to have the more acidic hydrogen by molecular
simulations. Steric hindrance to metalate C3’ position was also
invoked to account for the observed regioselectivity. When the
reaction is warmed up, equilibrations between C3’ and C5’
lithiated pyridines 28 and 29 are shifted towards the 3’-lithiated (-
50 °C). The amide group acts as an irreversible in situ trap for the
3’-lithiated pyridine to form the respective azafluorenone 27 along
with C5’ deuterated product after deuteriolysis. Through
thermodynamic control, the carbonyl group irreversibly traps the
metalated species at C3’ which shifts equilibrium towards the
formation of cyclized product 27.
The reaction mechanism of this transformation is significantly
different from the one described for the biphenyl-2-carboxylic
acid and carboxamide described earlier (see Schemes 3-6). We
propose the terminology “Directed isomerization of preformed
remote anions” to describe these transformation where the initial
metalation is independent from any remote unit.
Scheme 10. Directed isomerization of remote anion 29 (Mongin et al., 2004)[54]
Migrating DMG: Remote Anionic Fries Rearrangement
Electrophilic DMG with migrating abilities can translocate onto
metalated sites thereby reacting via remote anionic Fries
rearrangements (Scheme 11).[25, 45, 47-48, 53] On carbamate 2-
substituted biaryl structures, carbamoyl units migrate to the
9
kinetically favored metalation sites. The ortho positions
systematically need protection to enable remote anionic Fries
rearrangement. Remote metalation of these structures with
lithium amides follows an unfavorable equilibrium with the pKa
of amide bases often being lower than pKa of remote hydrogens.
Attempts to trap remote anions 30 with external electrophiles have
failed. A lack of measurable build-up of remote anionic species
indicates a rate-limiting remote metalation followed by a rapid
Fries rearrangement. The irreversible carbamoyl translocation
onto a remote anionic position likely drives the lithiation
equilibrium towards the rearranged Fries product 31 following Le
Châtelier’s principle.[53, 55] Additional activation of the remote
position with proximal substituents proves beneficial for control
of the regioselectivity of metalation. Tandem remote anionic Fries
followed by anionic Friedel-Crafts acylation were reported as a
one-pot process to form compounds 32.[23b, 23f, 26] DreM-DMG
translocation sequence was used in concert with DoM and cross
coupling chemistry to synthesize dibenzopyranones,[23c, 23i, 30, 53a]
defucogilvocarcin V[27],[26] and plicadin.[29] Treatment of N-
carbamoylindoles 34 with LDA provided a general route to
functionalized 2-arylindoles, benzo[a]carbazoles and indeno[1,2-
b]indoles.[48]
Scheme 11. Remote Anionic Fries rearrangement (Snieckus et al., 1994, 2008)[53]
Remote Tolyl Groups: Remote Lateral Metalations-
Nucleophilic Cyclizations
Directed remote lateral metalations with nucleophilic cyclizations
are reported to occur on flexible biaryl substrates bearing tolyl
groups at position remote from electrophilic DMG. An early
example was reported by Fouche and Leger: N-alkyl-N-(o-
tolyl)anthranilic esters 37 react with lithium diethylamide to form
dihydro-10,11-dibenzo[b,f]azepines 38 (Scheme 12).[52] Directed
remote lateral metalation and annulation was later described for
various systems using oximes ethers, hydrazones, nitriles, amides
as directing groups,[27] and were successfully applied to the
synthesis of 9-aminophenanthrenes,[27, 31] 9-phenanthrols,[56]
antiepileptic drug trileptal,[44b] eupolauramine[31],[30] β-
lapachone,[28] among other examples.[23g, 32, 53a] The pKa of amide
bases is lower than pKa of 4-methylbiphenyl methyl hydrogen
(>40.2).[57] Metalation of the remote lateral position was initially
explained through CIPE, and is likely favored by ISQ.[58]
Scheme 12. Directed lateral metalation-nucleophilic cyclizations on N-bridged
biarylic structures (Fouche and Leger, 1972)[52]
When competition between remote aromatic metalation and
remote lateral metalation is possible (Scheme 13), the outcome is
difficult to predict.[44]
Scheme 13. Competition between remote aromatic metalation-nucleophilic
cyclization and remote lateral metalation-nucleophilic cyclization.
Diarylamides substituted at position 2 by electrophilic DMGs
usually undergo lateral remote metalation-cyclization reactions in
the presence of lithium bases to form dibenzoazepinones. For
example, treatment of diarylamides 39, 40 and 41 with 2-4 equiv.
of LDA in THF at 0 °C yields dibenzoazepinone 42 preferentially
with good selectivities (Scheme 14). The selectivity depends on
the steric hindrance at the DMG. The activation of the remote
aromatic ring position with proximal substituents (OMe, Cl) can
sometimes allow a remote ring metalation-cyclization process to
compete to give acridones. In such cases, tuning of reaction
conditions is necessary to achieve selective transformations.[24, 44a]
Scheme 14. Directed Lateral remote metalation on N-[23h, 44a, 52] and O-bridged[24, 44a]
biarylamides (Snieckus et al. 1997, 2008)
10
In contrast to the good regioselectivity observed for product
formation in the dibenzo[b,f]azepinone series, LDA-mediated
cyclization of diphenyl ether 44 resulted in a reproducible
equimolar mixture of dibenzo[b,f ]oxepinone 45 and xanthone 46
(0 °C to rt, Scheme 14). Treatment of 44 with LDA at -78 °C
followed by warming to room temperature showed some
selectivity for formation of dibenzo[b,f]oxepinones (45/46 ≈ 2).
The oxygen bridge of diarylether 44 inductively activates remote
ring positions.[59] Reduced temperatures are required in order to
achieve lateral metalation on such reactants. Aromatic ring
metalations are favored over lateral metalation through a
collaborative effect of a second proximal activating group.[24, 44a]
To gain insights into the driving forces for those
transformations, DFT calculations were used to help rationalize
the selectivities of formation of cyclized products by remote
lateral metalation-nucleophilic cyclization for substrates 40 and
44.[44a] The lateral metalation products, 42 and 45, were calculated
to be thermodynamically unfavourable as these were less stable
than 43 and 46. The significance of other results obtained by DFT
calculations of low energy conformation of intermediate structures
leading to the cyclized products was unclear. The data seemed to
indicate that the rate constant of cyclization would likely
supersede conformation criteria during the reactions.
Optional Site Selectivity in DreM
When different regioisomers of a product can be formed, the
choice of DMG is critical for regioselective reactivity. For some
systems, it has been possible to switch regioselectivities of remote
metalation by changing DMGs.
2-(Pyridin-2-yl) and 2-(pyridin-4-yl) benzoic acids 47 and 49
afford azafluorenones 5H-indeno[1,2-b]pyridin-5-one 48 and 9H-
indeno[2,1-c]-pyridin-9-one 50 by treatment with LDA or LTMP
in THF at RT (Scheme 15).[60] The formation of azafluorenones
likely follows a mechanism similar to DreM of 2-
biphenylcarboxamide 12 (see Scheme 6).[36]
Scheme 15. DreM-cyclization on 2-(pyridin-2-yl) and 2-(pyridin-4-yl) benzoic acids
47 and 49 (Mongin et al., 2003) [60]
A dramatic change of regioselectivity occurs during the
DreM-cyclization of 2-(pyridin-3-yl)benzoic acid derivatives 51a-
c with lithium bases. Formation of either of the two azafluorenone
isomers with complete regiocontrol can be achieved by altering
the choice of DMG (Scheme 16).[35b]
Scheme 16. DreM-cyclization of 2-(pyridin-3-yl)benzoic acid derivatives 51a-c
(Mortier et al., 2006)[35b]
In 2-(pyridin-3-yl)benzoic acid derivatives 51a-c, the pyridine
nitrogen pyridine and Lewis basic DMG compete for coordination
with the lithium base. The DMG and the pyridine nitrogen are too
far apart to ensure double complexation with the base.[61] As the
DMG changes from CO2Et to CO2Li and CONEt2, the
coordinating properties of the DMG are altered. This potentially
modifies the preferential coordination site for lithium bases and
also affects the activation of remote hydrogens through both
complexation and electronic effects. Ester-LDA complexes are
unlikely,[15b, 15c, 62] thus coordination of the base to the pyridine
nitrogen is likely for 2-(pyridin-3-yl)benzoic ester 51c. The
preferential coordination site for the two other reactants is unclear.
In each case, the carbonyl derived DMG is expected to direct
metalations at ortho position and/or at C2’ and C4’. Potential
cumulative effects of nitrogen and carbonyl directing groups may
explain metalations of remote positions. Importantly, the DMGs
have varying electrophilicity. Thus switching DMGs strongly
affects the kinetics of the critical step of nucleophilic cyclization.
When DMG is CO2Et (strong electrophile, 51c), complexation
of the base with the nitrogen of the pyridine likely directs the
formation of the kinetic anion 52c which is destabilized by
electronic repulsion between the carbanion and the lone-pair of the
azine nitrogen. Species 52c does not have a long-enough lifetime
to isomerize to the thermodynamically more stable (less basic) 4’-
pyridyllithium 53c and therefore cyclizes rapidly to give
monolithium salt 54c leading to azafluorenone 56 as a sole
product after hydrolysis.
CONi-Pr2, a stronger Lewis base, competes with the pyridine
nitrogen for coordination. It also is a weak electrophile in
comparison to an ester. The azafluorenone 57 is obtained with
complete regiocontrol and high yield. Isomerization of a
potentially formed kinetic C2’ pyridine anion 52b to a more
thermodynamically stable C4’ anion 53b may take place before
cyclization (kinetically slower cyclization). Complexation of the
base to the amide followed by metalation at the most acidic
remote hydrogen position is also possible.[63] Carboxylate 51a
displays a reactivity pattern intermediate between the two previous
limiting cases. From compound 51a, both azafluorenones 56 and
57 are formed. Cyclizations of the remote carbanions 52 and 53
give cyclic salts 54 and 55 that are more thermodynamically stable
than an ortho carbanion.
11
DreM followed by nucleophilic cyclizations of benzoic acid
derivatives 51 are dictated through a combination of factors such
as: the acidity of the remote hydrogens, the ability of the directing
group to interact with the lithium base towards the formation of a
favorable low-energy transition state, activating remote metalation,
electrophilicity of DMG, and thermodynamic versus kinetic
considerations. None of these parameters should be considered
independently. Notably, no DreM is observed when carbonyl
derived directing groups are replaced by non-electrophilic OMe
units on structures 47, 49 and 51 (Scheme 17). DoM at the anisole
ring and/or at the pyridine ring occurs on 58 without further
isomerization.[64]
Scheme 17. Lithiation of anisylpyridines (Fort et al., 2005)[64]
DreM with Quench by External Electrophiles
Quench of remote anions with external electrophiles has been
reported for non-electrophilic coordinating DMG, and for
structurally rigid substrates that do not allow for intramolecular
quench by remote electrophilic non-migrating DMGs (Figure 2, C
and D). Few examples of quench of anions remote from DMGs
using external electrophiles are reported. Presumably, hydrogens
at positions remote from DMGs are less activated through
inductive effects and any driving force brought by intramolecular
quench of remote anion with electrophilic DMGs (ISQ technique)
is absent, thus making abstraction of remote hydrogens
challenging. This is in accordance with a nucleophilic cyclization
step being a significant driving force in some DreM-
intramolecular quench sequences. The use of alkyl lithium bases
that favour the formation of clusters is the norm for DreM using
non-electrophilic DMGs. To date, no mild conditions (i.e.-78 °C)
have been reported to generate remote anions quenchable by
external electrophiles. The following sections unravel the
mechanisms and driving forces of DreM followed by quench with
external electrophiles.
DreM on Rigid Substrates with Electrophilic Non-Migrating
DMG
When electrophilic coordinating and non-migrating DMGs are
present on rigid substrates, the lack of flexibility does not allow
for intramolecular quench of remote anions by DMG. In such
cases, quench by external electrophiles of remote metalated
species becomes possible.
Regioselective directed metalation of dibenzodioxin-1-
carboxylic acid 60 at the C9 position was achieved by treatment of
t-butyllithium (2 equiv. at -78 °C, Scheme 18).[65] Trapping of the
metalated species with various electrophiles afforded good yields
of 9-substituted products 61 only. Amide bases were not effective
in these transformations. Dibenzodioxin itself is not metalated at -
78 °C. Activation and remote direction of metalation at C9 by the
initially formed ArCO2Li group was confirmed with experimental
evidence.
Scheme 18. DreM of dibenzodioxin-1-carboxylic acid 60 (Palmer et al., 1990)[65]
Remote metalation from a carboxylate unit was also observed
on the related phenoxathiin ring system 62 by treatment with t-
butyllithium at -78 °C (Scheme 19).[65] Under the same reaction
conditions, the isomeric 1-carboxylic acid 64 failed to react
yielding only recovered starting material, presumably due to the
reduced ability of sulphur compared with oxygen to activate its
ortho position. Mechanistically, both the inductive acidifying
effect of the oxygen bridge and the anchoring effect of the
carboxylate group work in concert to bring about the regiospecific
remote metalation.
Scheme 19. Lithiations of phenoxathiin-4-carboxylic acid 62 and phenoxathiin-1-
carboxylic acid 64 (Palmer et al., 1990)[65]
Can Addition of External Electrophiles Initiate Remote
Metalation?
Much is known about the structures of lithium alkyls and amides
but less is known about how they react and the nature of the
lithiated species produced prior to workup of metalations. In the
case of directed metalation, lithiated species are usually not
characterized directly because of their high reactivity and
instability and their expected aggregation and solvation making
direct analysis complicated. Anions from metalation are trapped
with electrophiles and the efficiency of formation of quenched
products is used as indirect evidence to extrapolate prior
efficiency of regioselective metalation. Such interpretation can
sometimes be sometimes misleading. For example, it does not
apply to reactions using ISQ techniques, where high yields of
substituted products are obtained even though only trace amounts
of metalation occur.
The reaction of dibenzodioxin 65 at -30 °C with t-butyllithium
(2 equiv.) followed by quench with CO2 and esterification gives
the 1,9-diester from 61a (69%). This suggests that a 1,9-dianion
might be generated (Scheme 20).[65] However, when quenched
with iodomethane or dimethylformamide, only 1-substituted
dibenzodioxins 68 are isolated, indicating that at this temperature
only monometalation occurs to any appreciable extent.
Apparently, in the special case of quenching with CO2 the initially
formed 1-carboxylate salt is able to direct a subsequent remote
metalation at the 9-position during the quench. Mechanistically,
the kinetics of competitive reactions between the electrophiles and
12
the metalated species and mixed aggregates during the addition
could lead to formation of the DreM products. The electrophile
could have faster reaction kinetics with mixed aggregates than
with nBuLi in solution. Thus, remote metalation would be
initiated by addition of the electrophile. Analogous metalations
initiated by addition of electrophiles were reported on other
structures. [66]
Scheme 20. Electrophile mediated metalation of dibenzodioxin (65) with tBuLi.
(Palmer et al., 1990)[65]
DreM with a Non-Electrophilic Amino DMG
DreM of biarylamines assisted by an amino group were first
reported in the 1960s (Scheme 21).[19] In the presence of 4.5
equivalents of n-butyllithium at room temperature in diethylether,
metalations of 2-aminobiphenyls 69 at C2’ on the phenyl ring
followed by quench with dimethylformamide or carbon dioxide
led to the formation of cyclic products in moderate yields. [19] This
methodology was applied to the synthesis of phenanthridines 71
and phenanthridones 72. The syntheses of
dibenzo[b,f][1,4]oxazepine 75a and dibenzo[b,f][1,4]thiazepine
75b were described with C,N-dilithiation of 2-biarylamines being
the key step. Notably when X = NH, 1-phenylbenzimidazole 76
was formed by N-formylation instead of C-formylation.[67] Recent
improvements have been made in the metalation of metalate
substrates 73 at remote positions using 2 equivalents of n-
butyllithium/TMEDA in hexane (Scheme 21).[68] Under such
conditions, lithiation of 2-amino diphenylether 73a proceeds
rapidly with high regioselectivity. Upon quenching with excess
iodomethane, C,N-dimethylated product 78a is formed in 93%
yield (Scheme 21).[68a] Similarly, the di-lithiation of o-amino
diphenylsulfone 73d in THF affords the C,N-dimethylated product
78b. Other examples include the formation of 11-pyridyl-
dibenzo[b,f][1,4]oxazepines 77a, 11-(3-dimethylaminopropyl)-and
dibenzo[b,f][1,4]oxazepines 77b,c.[68b]The C,N-dilithio
intermediate 74 from o-amino diphenylsulfone affords the
tricyclic structure 77c after quench with methyl
cyclopropanecarboxylate. Mechanistically, CIPE and an increased
thermodynamic acidity of remote aromatic hydrogens for bridged
arenes likely assist these metalations.
Scheme 21. DreM of 2-aminobiaryl structures (Narasimhan et al., 1969, 1975,
1981)[19, 67] (Ogle et al., 1997, 1998) [68]
Further mechanistic insights to account for DreM directed by
the amino group were obtained by a combination of NMR and X-
ray analyses.[69] N,N-dilithiobiphenylamine crystallizes as a
cluster containing five RNLi2 units and NMR studies show
aggregations in solution. Alkyllithium bases, prone to forming
clusters, are necessary to achieve DreM. Milder conditions using
lithium amides proved inefficient to metalate remote positions
remote. Coordination of an alkyl lithium to the amino anchor
acting as ligand may preclude a subsequent remote metalation
analogously to the manner in which an amino group directs
exclusive perilithiation on 1-aminonaphtalene and on various
dibenzophenothiazines with no trace of ortho metalation.[7] The
regioselectivity of metalation may be linked to the thermodynamic
stability of the anions and aggregates formed.[70]
Isomerization of Preformed Anions Controlled by Remote
Non Electrophilic Substituents
Isomerization of preformed anionic species assisted by a Lewis
basic remote unit is included under DreM terminology. A
substituent directs the isomerization of a preformed carbanion by
stabilizing one of the remote metalated isomers through
complexation. The DMG does not cause the initial lithiation. We
propose the terminology: “directed isomerization of remote anion”,
to describe this type of transformation. This reactivity was
observed with 2,2-dimethyl-N-pyridylphenylpropanamides
(Schemes 22 and 23).[71] Lithiation of 2,2-dimethyl-N-[2-(2-
pyridyl)phenyl]propanamide 79 occurs at C6’ with in situ quench
by LTMP/SiMe3Cl (Scheme 22). Complexation of the lithium
base to the pyridine nitrogen most likely favors a kinetic
mechanism over an acid-base mechanism which would promote
the abstraction of the most acidic H4’ hydrogen. [71] When
lithiation of 79 with LTMP was quenched with a deuterium source
at room temperature, a mixture of deuterated isomers 81, 82, 83
was observed. Under reversible lithiation conditions,
isomerization of the initially-formed 6-lithiopyridine to more
stable derivatives at C3’ (stabilization through complexation of the
metal with the DMG at remote position) and at C4’ or C5’
13
(reduction of electronic repulsion between the carbanion and the
lone pair of the azine nitrogen) was observed.
Scheme 22. Metalation of of 2,2-dimethyl-N-[2-(2-pyridyl)phenyl]propanamide (79)
(Queguiner et al., 2003)[71]
Directed remote isomerization is more apparent on 2,2-
dimethyl-N-[2-(3-pyridyl)phenyl]propanamide 84 (Scheme 23).
Metalation occurs at C6’ and C2’ using in situ quenching with
LTMP/Me3SiCl (85/86, 66:34). Complexation of the lithium to
the pyridine nitrogen probably favors a kinetic mechanism. When
deuteriolysis was conducted after 2 hours at 0 °C, only deuterated
product at C4’ was isolated (87, 50% deuterium incorporation,
70% yield). Under thermodynamic control using reversible
lithiation conditions, isomerization of the initially formed
lithiopyridines 88 and 91 to the more stable derivative 90
(stabilization through complexation of the metal with the remote
DMG) through a series of equilibria is observed.
Scheme 23. Metalations of 2,2-dimethyl-N-[2-(3-pyridyl)phenyl]propanamide (84)
(Queguiner et al., 2003) [71]
Summary and Outlook
The synthetic utility of directed remote aromatic metalation
(DreM) is an outcome of good reaction yields and unique
reactivity that can be complementarity to Friedel-Crafts
transformations. The increasingly widespread use of DreM is, in
large part, a result of extensive development efforts in the
Snieckus laboratory. CIPE can be considered a useful predicative
model but, given the examples presented above, it should not be
considered a complete and systematic explanation for directed
metalation reactivity without additional mechanistic investigation.
Detailed mechanistic studies of organolithium reactions as
discussed in this review can contribute to identify potentially-
misleading models for rationalizing regioselectivity and product
distributions. Further insight into DreM mechanisms with
increased accuracy will continue allow for new regioselective and
efficient syntheses. Continued development and will enable a
foundation for intelligent design of new transformations just as the
DoM methodology did for C-H bond activations using transition
metals.[72]
[1] H. W. Gschwend, H. R. Rodriguez, Org. React. 1979, 26, 1.
[2] a) V. Snieckus, Chem. Rev. 1990, 90, 879-933; b) M. C. Whisler, S. MacNeil,
V. Snieckus, P. Beak, Angew. Chem. Int. Ed. 2004, 43, 2206-2225.
[3] a) P. Beak, A. Basu, D. J. Gallagher, Y. S. Park, S. Thayumanavan, Acc.
Chem. Res. 1996, 29, 552-560; b) P. Beak, A. I. Meyers, Acc. Chem. Res.
1986, 19, 356-363.
[4] a) M. Schlosser, Organometallics in Synthesis: A Manual, Wiley, Chichester,
1994; b) M. Schlosser, Angew. Chem. Int. Ed. 2005, 44, 376-393.
[5] J. Clayden, Organolithiums: Selectivity for Synthesis, Pergamon, Oxford,
2002.
[6] T. Klis, S. Lulinski, J. Serwatowski, Curr. Org. Chem. 2008, 12, 1479-1501.
[7] a) J. Clayden, C. S. Frampton, C. McCarthy, N. Westlund, Tetrahedron 1999,
55, 14161-14184; b) R. A. Barnes, L. J. Nehmsmann, J. Org. Chem. 1962, 27,
1939-1942; c) A. J. Kirby, J. M. Percy, Tetrahedron 1988, 44, 6903-6910.
[8] G. W. Gribble, in Name Reactions for Homologations (Ed.: J. J. Li), John
Wiley & Sons, Hoboken, N.J., 2009, pp. 749-764.
[9] a) T. H. Nguyen, N. T. T. Chau, A. S. Castanet, K. P. P. Nguyen, J. Mortier,
Org. Lett. 2005, 7, 2445-2448; b) D. R. Anderson, N. C. Faibish, P. Beak, J.
Am. Chem. Soc. 1999, 121, 7553-7558.
[10] a) D. J. Gallagher, H. Du, S. A. Long, P. Beak, J. Am. Chem. Soc. 1996, 118,
11391-11398; b) A. Streitwieser, J. E. Williams, S. Alexandratos, J. M.
Mckelvey, J. Am. Chem. Soc. 1976, 98, 4778-4784.
[11] a) D. J. Gallagher, P. Beak, J. Org. Chem. 1995, 60, 7092-7093; b) H. Luitjes,
F. J. J. Dekanter, M. Schakel, R. F. Schmitz, G. W. Klumpp, J. Am. Chem.
Soc. 1995, 117, 4179-4180; c) D. J. Pippel, G. A. Weisenburger, N. C. Faibish,
P. Beak, J. Am. Chem. Soc. 2001, 123, 4919-4927; d) S. T. Chadwick, R. A.
Rennels, J. L. Rutherford, D. B. Collum, J. Am. Chem. Soc. 2000, 122, 8640-
8647.
[12] D. R. Hay, Z. G. Song, S. G. Smith, P. Beak, J. Am. Chem. Soc. 1988, 110,
8145-8153.
[13] For recent investigations supporting the formation of complexes on the
pathway of directed metalations of some reactants, see: a) F. Affortunato, S.
Florio, R. Luisi, B. Musio, J. Org. Chem. 2008, 73, 9214-9220; b) C.
Strohmann, V. H. Gessner, Chem. Asian J. 2008, 3, 1929-1934; c) I.
Fernandez, J. Gonzalez, F. Lopez-Ortiz, J. Am. Chem. Soc. 2004, 126, 12551-
12564.
[14] a) J. C. Riggs, K. J. Singh, M. Yun, D. B. Collum, J. Am. Chem. Soc. 2008,
130, 13709-13717; b) K. J. Singh, D. B. Collum, J. Am. Chem. Soc. 2006, 128,
13753-13760; c) S. T. Chadwick, A. Ramirez, L. Gupta, D. B. Collum, J. Am.
Chem. Soc. 2007, 129, 2259-2268.
[15] a) D. B. Collum, Accounts Chem. Res. 1992, 25, 448-454; b) X. F. Sun, D. B.
Collum, J. Am. Chem. Soc. 2000, 122, 2459-2463; c) X. F. Sun, D. B. Collum,
J. Am. Chem. Soc. 2000, 122, 2452-2458.
[16] N. J. R. V. Hommes, P. V. Schleyer, Angew. Chem. Int. Ed. 1992, 31, 755-
758.
[17] D. W. Slocum, A. Carroll, P. Dietzel, S. Eilerman, J. P. Culver, B. McClure, S.
Brown, R. W. Holman, Tetrahedron Lett. 2006, 47, 865-868.
[18] a) R. Maggi, M. Schlosser, Tetrahedron Lett. 1999, 40, 8797-8800; b) M.
Schlosser, F. Mongin, J. Porwisiak, W. Dmowski, H. H. Buker, N. M. M.
Nibbering, Chem. Eur. J. 1998, 4, 1281-1286; c) M. Schlosser, Angew. Chem.
Int. Ed. 1998, 37, 1497-1513; d) T. H. Nguyen, N. T. T. Chau, A. S. Castanet,
K. P. P. Nguyen, J. Mortier, J. Org. Chem. 2007, 72, 3419-3429.
[19] a) Narasimh.Ns, R. H. Alurkar, Indian. J. Chem. 1969, 7, 1280; b) N. S.
Narasimhan, P. S. Chandrachood, Synthesis 1979, 589-590; c) N. S.
Narasimhan, R. S. Mali, Tetrahedron 1975, 31, 1005-1009; d) N. S.
Narasimhan, P. S. Chandrachood, N. R. Shete, Tetrahedron 1981, 37, 825-
827.
[20] M. E. K. Cartoon, G. W. H. Cheeseman, J. Organomet. Chem. 1981, 212, 1-9.
14
[21] G. W. H. Cheeseman, S. G. Greenberg, J. Organomet. Chem. 1979, 166, 139-
152.
[22] J. M. Fu, B. P. Zhao, M. J. Sharp, V. Snieckus, J. Org. Chem. 1991, 56, 1683-
1685.
[23] For reviews on application of DreM in synthesis see: a) C. G. Hartung, V.
Snieckus, (Ed.: D. Astruc), Wiley-VCH, New York, 2002, p. 330; b) J. M. Fu,
B. P. Zhao, M. J. Sharp, V. Snieckus, Can. J. Chem. 1994, 72, 227-236; c) B.
Chauder, L. Green, V. Snieckus, Pure Appl. Chem. 1999, 71, 1521-1529; d) L.
Green, B. Chauder, V. Snieckus, J. Heterocyclic. Chem. 1999, 36, 1453-1468;
For leading references see: e) S. L. MacNeil, B. J. Wilson, V. Snieckus, Org.
Lett. 2006, 8, 1133-1136; f) C. A. James, V. Snieckus, J. Org. Chem. 2009, 74,
4080-4093; g) X. W. Cai, V. Snieckus, Org. Lett. 2004, 6, 2293-2295; h) N.
Thasana, S. Ruchirawat, Synlett 2003, 1037-1039; i) B. A. Chauder, A. V.
Kalinin, N. J. Taylor, V. Snieckus, Angew. Chem. Int. Ed. 1999, 38, 1435-
1438.
[24] O. B. Familoni, I. Ionica, J. F. Bower, V. Snieckus, Synlett 1997, 1081-1083.
[25] W. Wang, V. Snieckus, J. Org. Chem. 1992, 57, 424-426.
[26] C. A. James, V. Snieckus, Tetrahedron Lett. 1997, 38, 8149-8152.
[27] L. Benesch, P. Bury, D. Guillaneux, S. Houldsworth, X. Wang, V. Snieckus,
Tetrahedron Lett. 1998, 39, 961-964.
[28] M. A. F. Brandao, A. B. Deoliveira, V. Snieckus, Tetrahedron Lett. 1993, 34,
2437-2440.
[29] B. A. Chauder, A. V. Kalinin, N. J. Taylor, V. Snieckus, Angew. Chem. 1999,
111, 1528.
[30] X. Wang, V. Snieckus, Tetrahedron Lett. 1991, 32, 4883-4884.
[31] X. Wang, V. Snieckus, Tetrahedron Lett. 1991, 32, 4879-4882.
[32] J. Board, J. X. Wang, A. P. Crew, M. Z. Jin, K. Foreman, M. J. Mulvihill, V.
Snieckus, Org. Lett. 2009, 11, 5118-5121.
[33] X. Wang, PhD Thesis, University of Waterloo 1992.
[34] a) M. J. Sharp, W. Cheng, V. Snieckus, Tetrahedron Lett. 1987, 28, 5093-
5096; b) W. Cheng, V. Snieckus, Tetrahedron Lett. 1987, 28, 5097-5098; c) D.
Tilly, S. S. Samanta, F. Faigl, J. Mortier, Tetrahedron Lett. 2002, 43, 8347-
8350; d) D. Tilly, S. S. Samanta, A. S. Castanet, A. De, J. Mortier, Eur. J.
Org. Chem. 2005, 174-182.
[35] a) D. Tilly, S. S. Samanta, A. De, A. S. Castanet, J. Mortier, Org. Lett. 2005,
7, 827-830; b) D. Tilly, A. S. Castanet, J. Mortier, Tetrahedron Lett. 2006, 47,
1121-1123.
[36] D. Tilly, J. M. Fu, B. P. Zhao, M. Alessi, A. S. Castanet, V. Snieckus, J.
Mortier, Org. Lett. 2010, 12, 68-71.
[37] Some electrophiles are compatible with lithium bases at low temperatures
(such as Me3SiCl, Me2Cl2Si, B(OiPr)3, B(OMe)3). When lithiation is not
complete at the time of addition of the electrophile to the reaction mixture,
reaction of the anion with the electrophile is faster than the reaction of the
electrophile with the metalating agent. The equilibria are shifted towards the
formation of substituted product through kinetic control. For examples see: a)
T. D. Krizan, J. C. Martin, J. Am. Chem. Soc. 1983, 105, 6155-6157; b) S.
Caron, J. M. Hawkins, J. Org. Chem. 1998, 63, 2054-2055; c) B. H. Lipshutz,
M. R. Wood, C. W. Lindsley, Tetrahedron Lett. 1995, 36, 4385-4388.
[38] For properties of LiCKOR see: a) M. Schlosser, Mod. Syn. Met. 1992, 6, 227-
271; b) M. Schlosser, Pure Appl. Chem. 1988, 60, 1627-1634; c) M. Schlosser,
J. H. Choi, S. Takagishi, Tetrahedron 1990, 46, 5633-5648; d) L. Lochmann,
Collect. Czech. Chem. Commun. 1987, 52, 2710-2716.
[39] M. E. K. Cartoon, G. W. H. Cheeseman, J. Organomet. Chem. 1982, 234,
123-136.
[40] M. Stratakis, J. Org. Chem. 1997, 62, 3024-3025.
[41] H. F. Bluhm, H. V. Donn, H. D. Zook, J. Am. Chem. Soc. 1955, 77, 4406-
4407.
[42] For hidden proton effect see: a) D. Seebach, Angew. Chem. Int. Ed. 1988, 27,
1624-1654; b) T. Weber, D. Seebach, Helv. Chim. Acta. 1985, 68, 155-161; c)
T. Laube, J. D. Dunitz, D. Seebach, Helv. Chim. Acta. 1985, 68, 1373-1393.
[43] M. Alessi, A. L. Larkin, K. A. Ogilvie, L. A. Green, S. Lai, S. Lopez, V.
Snieckus, J. Org. Chem. 2007, 72, 1588-1594.
[44] a) S. L. MacNeil, M. Gray, D. G. Gusev, L. E. Briggs, V. Snieckus, J. Org.
Chem. 2008, 73, 9710-9719; b) O. Lohse, U. Beutler, P. Funfschilling, P.
Furet, J. France, D. Kaufmann, G. Penn, W. Zaugg, Tetrahedron Lett. 2001,
42, 385-389.
[45] S. I. Mohri, M. Stefinovic, V. Snieckus, J. Org. Chem. 1997, 62, 7072-7073.
[46] a) R. R. Fraser, M. Bresse, T. S. Mansour, J. Am. Chem. Soc. 1983, 105,
7790-7791; For relative acidity of hydrogens on heterocycles see: b) O. A.
Reutov, I. P. Beletskaya, K. P. Butin, CH Acids, Pergamon Press, Oxford,
1978.
[47] C. A. James, A. L. Coelho, M. Gevaert, P. Forgione, V. Snieckus, J. Org.
Chem. 2009, 74, 4094-4103.
[48] M. Gray, B. J. Chapell, N. J. Taylor, V. Snieckus, Angew. Chem. Int. Ed.
1996, 35, 1558-1560.
[49] a) A. S. Rebstock, F. Mongin, F. Trecourt, G. Queguiner, Org. Biomol. Chem.
2004, 2, 291-295; b) J. Epsztajn, A. Jozwiak, J. K. Krysiak, D. Lucka,
Tetrahedron 1996, 52, 11025-11036.
[50] G. Queguiner, F. Marsais, V. Snieckus, J. Epsztajn, Adv. Heterocycl. Chem.
1991, 52, 187-304.
[51] A. S. Castanet, D. Tilly, J. B. Veron, S. S. Samanta, A. De, T. Ganguly, J.
Mortier, Tetrahedron 2008, 64, 3331-3336.
[52] a) J. Fouche, A. Leger, GermanPatent 2,202,486, 1972; Chem. Abstr. 1972, 77,
152012; b) J. Fouche, A. Leger, GermanPatent 2,239,396, 1972; Chem. Abstr.
1972, 77, 152012
[53] For remote remote anionic Fries reactivity see: a) S. L. MacNeil, M. Gray, L.
E. Briggs, J. J. Li, V. Snieckus, Synlett 1998, 419; b) Z. Zhao, A. Jaworski, I.
Piel, V. Snieckus, Org. Lett. 2008, 10, 2617-2620; c) F. Beaulieu, V. Snieckus,
J. Org. Chem. 1994, 59, 6508-6509; d) T. K. Macklin, M. A. Reed, V.
Snieckus, Eur. J. Org. Chem. 2008, 1507-1509.
[54] A. S. Rebstock, F. Mongin, F. Trecourt, G. Queguiner, Tetrahedron 2004, 60,
2181-2186.
[55] D. Tilly, PhD Thesis, University of Maine, 2004.
[56] X. W. Cai, S. Brown, P. Hodson, V. Snieckus, Can. J. Chem. 2004, 82, 195-
205.
[57] A. Streitwieser, F. Guibe, J. Am. Chem. Soc. 1978, 100, 4532-4534.
[58] J. M. Fu, V. Snieckus, Can. J. Chem. 2000, 78, 905-919.
[59] For dilithiation of dibenzofuran see: a) H. D. Wang, F. P. Gabbai,
Organometallics 2005, 24, 2898-2902; b) K. Y. Tsang, H. Diaz, N. Graciani,
J. W. Kelly, J. Am. Chem. Soc. 1994, 116, 3988-4005; c) H. Gilman, W. J.
Trepka, J. Org. Chem. 1961, 26, 5202; d) I. Granoth, C. Symmes, J. B. Levy,
J. Chem. Soc. Perk. Trans. II 1972, 697.
[60] A. S. Rebstock, F. Mongin, F. Trecourt, G. Queguiner, Tetrahedron 2003, 59,
4973-4977.
[61] P. Gros, S. Choppin, Y. Fort, J. Org. Chem. 2003, 68, 2243-2247.
[62] X. F. Sun, S. L. Kenkre, J. F. Remenar, J. H. Gilchrist, D. B. Collum, J. Am.
Chem. Soc. 1997, 119, 4765-4766.
[63] The transformation of amide 51b to 57 is a non published result by D. Tilly.
[64] M. L. Parmentier, P. Gros, Y. Fort, Tetrahedron 2005, 61, 3261-3269.
[65] B. D. Palmer, M. Boyd, W. A. Denny, J. Org. Chem. 1990, 55, 438-441.
[66] a) F. Gohier, A. S. Castanet, J. Mortier, J. Org. Chem. 2005, 70, 1501-1504;
b) J. Mortier, T. H. Nguyen, D. Tilly, A. S. Castanet, Arkivoc 2007, 47-54.
[67] L. A. Crawford, H. Mcnab, A. R. Mount, S. I. Wharton, J. Org. Chem. 2008,
73, 6642-6646.
[68] a) A. Chopra, D. C. Dorton, C. A. Ogle, Main Group Met. Chem. 1998, 21,
25-28; b) A. Chopra, D. C. Dorton, C. A. Ogle, Main Group Met. Chem. 1997,
20, 783-786.
[69] C. Neumann, A. Schulz, T. Seifert, W. Storch, M. Vosteen, Eur. J. Inorg.
Chem. 2002, 1040-1045.
[70] Participation of an electrophile initiated direct remote aromatic metalation to
explain DreM of biarylamines using DMF and CO2 cannot be ruled out yet.
Reactions of aminolithium with CO2 or DMF during the quenching step could
form a new DMG. N-carboxylate unit is known to activate subsequent C-
lithiation steps. For examples see: ) A. R. Katritzky, L. M. V. Demiguel, G.
W. Rewcastle, Synthesis 1988, 215-217; b) A. R. Katritzky, M. Black, W. Q.
Fan, J. Org. Chem. 1991, 56, 5045-5048; Directing metalation properties of
aminoalkoxides have already been described. For examples see: c) D. L.
Comins, J. D. Brown, J. Org. Chem. 1984, 49, 1078-1083; d) A. Hallberg, A.
Svensson, A. R. Martin, Tetrahedron Lett. 1986, 27, 1959-1962; N-
carboxylate or aminoalkoxide would direct a remote metalation, the remote
lithiated species could then be trapped (either through irreversible
intramolecular anionic N-Fries rearrangement or by dry ice/DMF
respectively) to afford the cyclized product.
[71] A. S. Rebstock, F. Mongin, F. Trecourt, G. Queguiner, Org. Biomol. Chem.
2003, 1, 3064-3068.
[72] X. Chen, K. M. Engle, D. H. Wang, J. Q. Yu, Angew. Chem. Int. Ed. 2009, 48,
5094-5115.
Received: ((will be filled in by the editorial staff)) Revised: ((will be filled in by the editorial staff))
Published online: ((will be filled in by the editorial staff))
15
[a]* Dr. D. Tilly Eskitis Institute for Cell and Molecular Therapies Griffith University Don Young Road, Nathan 4111, Queensland, Australia E-mail: dp.tilly@laposte.net
[b] Dr. J. Magolan Department of Chemistry, University of Idaho Moscow, ID, USA
[c] Prof. J. Mortier University of Le Mans UCO2M (UMR CNRS 6011) Avenue Olivier Messiaen, 72085 Le Mans, Cedex 9, France
16
Entry for the Table of Contents (Please choose one layout only)
Layout 1:
Long-distance directed metalation.
Directed remote aromatic metalation
is a valuable synthetic tool that has
been utilized for the efficient
preparation of elaborate aromatic
scaffolds. This review tackles the
mechanistic nuances and driving
forces behind remote metalation in
order to unravel the causes of the
observed regiochemical outcomes and
enable enhanced predictability of
these remarkable synthetic
transformations.
David Tilly*,[a]
Jakob Magolan,[b]
Jacques Mortier
Directed Remote Aromatic
Metalations: Mechanisms and
Driving Forces
Recommended